US 20060154608 A1 Abstract A pilot extract section
14 extracts a pilot signal from received signal. An adder 21 in-phase adds a plurality of correlated values of pilot signals for respective subcarriers. Delay devices 201-1 to 201-6 temporarily maintain one in-phase added value. Multiplier 202-1 to 202-6 multiplies a predetermined coefficient to the in-phase added value that is output from the delay device. The predetermined coefficient reflects the result, which is obtained by correcting for multiple times the difference of channel variation in the different subcarriers that is generated when noise power per one subcarrier is calculated. Each of the multiplying results is added by an adder 24, and is squared by a square device 25. A cumulative adder 26 cumulative-adds the squared values for the whole subcarrier. A multiplier 203 averages by multiplying predetermined values to the cumulative-added values. Having such configuration, even if the correlation between adjacent subcarriers becomes lower by frequency selective fading, the accuracy for estimating noise power can be improved. Claims(6) 1. A method for estimating noise power, comprising:
averaging correlated values of known signal arranged in a plurality of subcarrier waves; calculating noise power per one subcarrier wave using the averaged values of the correlated values of said known signal; correcting calculation error of said noise power generated by difference of channel variation between subcarrier waves for multiple times based on said known signal; cumulative-adding values of noise power per one subcarrier wave, which are corrected in desired subcarrier wave band; and averaging thereof by multiplying a predetermined value to the value obtained by the cumulative-adding operation to estimate noise power. 2. The method for estimating noise power according to correcting calculation error of said noise power per one subcarrier wave generated by difference of channel variation between subcarrier waves, using an average value of the correlated value of said known signal in the subcarrier wave in question; and further repeatedly correcting an error generated by said correction. 3. The method for estimating noise power according to 4. The method for estimating noise power according to measuring a level of spreading of the multipath based on the received signal; determining a level of the correlation between adjacent subcarrier waves by the level of the measured spreading of the multipath, wherein number of said employed plurality of adjacent subcarrier waves is increased when the correlation between adjacent subcarrier waves is high, and wherein number of said employed plurality of adjacent subcarrier waves is decreased when the correlation between adjacent subcarrier waves is low. 5. The method for estimating noise power according to estimating Doppler frequency based on the received signal, wherein number of employed known signal arranged along the time orientation is increased when estimated Doppler frequency is low, p 1 wherein number of employed known signal arranged along the time orientation is decreased when estimated Doppler frequency is high, and wherein the value obtained by in-phase adding the correlated value of a plurality of known signals in each of subcarrier waves is employed as a correlated value of said known signal. 6. A noise power estimation apparatus, comprising:
a known signal-extracting means for extracting known signal from signal transmitted by using a plurality of subcarrier waves from a communication partner; a first multiplying means for multiplying respective predetermined coefficients to correlated values of said known signal between a plurality of adjacent subcarrier waves; a square means for calculating noise power per one subcarrier wave by squaring after adding said multiplying result; a cumulative-adding means for cumulative-adding noise power per one subcarrier wave calculated by said square means for desired subcarrier waves; and a second multiplying means for multiplying a predetermined value to the cumulative-added value calculated by said cumulative-adding means to obtain an averaging value, wherein said first multiplying means has a predetermined coefficient reflecting a result, which is obtained by correcting for multiple times a calculation error of noise power per one subcarrier wave generated by a difference in a channel variation between subcarrier waves based on said known signal. Description The present invention relates to a method for estimating noise power and a noise power estimation apparatus, and is preferable for being applied to, for example, radio receiver apparatus. In multi carrier-code division multiple access (MC-CDMA), there is a method for spreading to a frequency domain. In this method, after the information data stream is spreaded with a given spreading code series, each chip is mapped to different subcarriers. In these packets shown in this figure, one frame (54 symbols) is composed of two slots. In one frame, P-PICH, D-PICH and DPCH are time division-arranged. As can be seen from the figure, four symbols are arranged in one frame for the D-PICH. Conventionally, in order to ideally conduct a control such as adaptive modulation or a resource allocation, it is necessary to appropriately measure the quality of received signal, and noise power maybe employed as a quality of received signal. The noise power includes a thermal noise generated inside a receiver and a noise generated by an interference of the other cell through the propagation path. The D-PICH having the above-described packet configuration is employed for the estimation of the noise power (referred to as “N+I Pilot extract section The noise power estimation section P/S converter The delay device The multipliers In the next, an arithmetic operation conducted within the noise power estimation section Alternatively, ξ We will specifically find out “N+I Here, ε The following formula (5) can be derived from formula (2) and formula (3):
In this case, correlations between adjacent subcarrier waves are not necessarily be 1 for subcarriers i−1 to i+1, and thus there may be a case that a difference in channel variation h is generated. In other words, since the value obtained by subtracting h Here, hˆ On the other hand, hˆ In this way, h can more precisely be found by conducting the averaging operation using ξ across a plurality of subcarriers, thereby diminishing an error of σi. Formula (7) and formula (8) are substituted into formula (6) to obtain formula (9) shown below:
Here, assuming that differences of respective channel variation components are equal, i.e., h Formula (11) is provided by conducting an approximating calculation of formula (9) using formula (10):
Formula (11) presents an arithmetic operation conducted in the square device In this formula, coefficients of respective ξ are the coefficients that are set by multipliers Assuming that α=(⅛) In this case, cumulative-adding operation for σ′ that is output from the square device Estimations of noise power has conventionally been carried out by the method stated above. However, in the above-described conventional process for estimating noise power, there is a problem, in which noise power cannot be precisely estimated when the channel variations in respective subcarriers are large by the influence of frequency selectivity fading or the like. An object of the present invention is to provide a method for estimating noise power and a noise power estimation apparatus, which provide improvement in the estimation accuracy of the noise power, even if the correlation between adjacent subcarriers is low due to the frequency selective fading. The above-described object is achieved by repeatedly correcting differences in the channel variations in different subcarriers for multiple times when noise power per one frame (“N+I Embodiments of the present invention will be described in reference with the annexed figures as follows. An adder Delay devices Predetermined coefficients are set in advance, and the multipliers The multiplier Next, an arithmetic operation conducted in a noise power estimation section stated above will be described using formula. However, since the processes of formula (1) to (9) are same as stated above, description will be made for operations after formula (9). In formula (9), a correction same as formula (6) is further introduced in order to diminish an error caused by channel variation component h. This is because, when a drift between the subcarriers is occurred due to frequency selective fading, correlation of the adjacent subcarriers is decreased to increase values obtained by subtracting the channel variation components generated between subcarriers therefrom, and thus the purpose for the correction is to estimate noise power (“N+I Expand formula (15) to obtain formula (16):
Here, assuming that differences of respective channel variation components are equal, i.e., h Approximating formula (16) by employing formula (17) provides the following formula (18):
Actually, the calculation of this formula (18) is carried out by replacing n with ξ.
Coefficients set for multipliers Assuming α=( 1/32) In the cumulative adder As such, according to the present embodiment, errors generated by the channel variations can be diminished by introducing corrections twice during the process for estimating noise power, and thus the estimation accuracy of the noise power can be achieved, even if the correlations between adjacent subcarriers are decreased due to the frequency selective fading. While the first embodiment illustrates the case of using ξ that are obtained by averaging ξ of adjacent two subcarriers when noise power per one subcarrier is calculated, the second embodiment will illustrate a case of using ξ that are obtained by averaging ξ of adjacent four subcarriers. The measurement results represent conditions of frequency selective fading, and levels of the correlations of the adjacent subcarriers can be judged by using thereof. More specifically, the delay spread determination section The noise power estimation section Arithmetic operations provided by the noise power estimation section Similarly as in the first embodiment, a difference value obtained by subtracting ε Formula (22) and formula (23) are substituted into formula (24) and the resultant formula is further simplified to obtain the following formula (25):
Here, a correction is added to formula (25) in order to diminish error generated by channel variation:
hˆ Formula (27) and formula (28) are substituted into formula (26) to obtain formula (29):
A correction is further introduced to formula (29), for the purpose of reducing an influence thereto by the channel variation components, in consideration of frequency selective fading.
Expand formula (30) to obtain the following formula (31):
Here, assuming that differences in respective channel variation components are equal, the following formula (32) is obtained:
Approximate formula (31) by employing formula (32) to obtain the following formula (33):
Actually, the calculation of this formula (33) is carried out by replacing n with ξ. Calculating operation for obtaining square thereof is conducted in the square device Coefficients set for multipliers Here, provided that ni−6 to ni+6 satisfy probability distribution of Gaussian distribution in formula (33) and average power is presented as (na) Assuming α=(1/256) In the cumulative adder As such, according to the present embodiment, errors generated by the channel variations can be diminished by introducing corrections twice during the process for estimating noise power, when ξ of adjacent four subcarriers are averaged in the calculation for obtaining noise power per one subcarrier, and thus the estimation accuracy of the noise power can be achieved, even if the correlation between adjacent subcarriers is decreased due to the frequency selective fading. While the first embodiment illustrates the case of twice conducting the correcting operations during the process for estimating noise power, the third embodiment will illustrate a case of conducting the correcting operations for three times during the process for estimating noise power. Arithmetic operations provided by the noise power estimation section shown in In the third embodiment, additional correction is further added to formula (16), and thereafter, similarly as each of the embodiments stated above, expansion and approximation are carried out to obtain the following formula (37):
Actually, the calculation of this formula (37) is carried out by replacing n with ξ.
Coefficients set for multipliers Here, assuming that α=(1/128) As such, according to the present embodiment, the estimation accuracy of the noise power can further be achieved, by adding corrections conducted for three times during the process for estimating noise power. While the third embodiment illustrates a case of conducting the correcting operations for three times during the process for estimating noise power, in the configuration of using ξ that are obtained by averaging ξ of adjacent two subcarriers when noise power of subcarrier is calculated, the fourth embodiment will illustrate a case of conducting the correcting operations for three times during the process for estimating noise power in the case of using ξ that are obtained by averaging ξ of adjacent four subcarriers. Arithmetic operations provided by the noise power estimation section shown in Here, specific arithmetic operations will be described focusing on a subcarrier i. However, since the operations until formula (31) is derived in this example is the same as in the second embodiment, description will be made for operations after obtaining formula (31). In the fourth embodiment, additional correction is further added to formula (31), and thereafter, similarly as each of the embodiments stated above, expansion and approximation are carried out to obtain the following formula (41):
Actually, the calculation of this formula (41) is carried out by replacing n with ξ.
Coefficients set for multipliers Here, provided that n Here, assuming that α=(1/2048) As such, according to the present embodiment, the estimation accuracy of the noise power can further be achieved, by conducting corrections for three times during the process for estimating noise , in the case of using ξ that are obtained by averaging ξ of adjacent four subcarriers when noise power of subcarrier is calculated. While the second embodiment illustrates a case of adjusting number of subcarrier employed in the averaging operation, or in other words, a case of adjusting number of pilots arranged along the frequency orientation, based on the measurement result of a delay spread, which is measured by estimating the condition of frequency selective fading, the fifth embodiment of the present invention will illustrate a case of adjusting number of pilots arranged along the time orientation employed in the averaging operation. The fD estimation section The noise power estimation section According to the present embodiment, number of PICH arranged along the time orientation employed in the in-phase adding operation is changed based on Doppler frequency, so that the influence to the fading is reduced, and the accuracy for estimating noise power is further improved. As have been described above, according to the present invention, the accuracy for estimating the noise power can be improved by repeatedly correcting for multiple times the differences in the channel variations in the different subcarriers when noise power is estimated, even if the correlation between adjacent subcarriers is low due to the frequency selective fading. Although each of the above preferred embodiments describes that corrected values of noise power per one subcarrier, which are obtained for the whole subcarrier wave, are cumulative-added, and the obtained values are averaged to estimate noise power, the present invention is not limited thereto, and corrected values of noise power per one subcarrier, which are obtained for a desired subcarrier, may also be cumulative-added, and the obtained values may be averaged by using the desired subcarrier. In addition, the multiplex system for pilot signal (PICH) described in each of the embodiments stated above may be any multiplex system, such as time multiplex, frequency multiplex, scattered multiplex, code multiplex and the like. In addition, the method for estimating noise power and the noise power estimation apparatus of the present invention may be applied to Multiple Input Multiple Output—Orthogonal Frequency Division Multiplexing (MIMO-OFDM) or transmission diversity. It is considered that, in the MIMO-OFDM or the transmission diversity, there are basically a lot of scatters, and the propagation paths for each of the antenna are influenced by highly-selective fading, and thus received power considerably fluctuates between subcarriers. Thus, the estimation accuracy of noise power can be improved by applying method for estimating noise power and the noise power estimation apparatus of the present invention. The present application is based on Japanese patent application No.2003-038935, filed Feb. 17, 2003, the whole contents of which are hereby incorporated by reference. The present invention is preferable to be applied to a radio receiver. Referenced by
Classifications
Legal Events
Rotate |